Every Company has a vision for the future. Realizing the vision requires a series of clear steps, a roadmap from current state to the future state. To traverse the road to the future, you must leverage your current core competencies to offset your shortcomings and develop business strategies by analyzing the industry framework, competitive landscape, the interrelation between supply chain processes and cost-benefit tradeoffs. To understand all of the above mentioned factors and realize the vision, we need to analyze the existing network to see what is working and what is not. Network Design Modeling helps in opening two locks with one key, you can come up with the strategies by analyzing the current state as well as the coordinate the supply chain processes to realize those strategies.
As much as the small details are necessary to execute the steps, they are not needed to arrive at the roadmap. Hundreds of customers and thousands of products make up the supply chain of an average retail company in the United States. Do we need to account for those big numbers in an individual level to arrive at a growth strategy? Do we need to consider ten products of similar characteristics individually to decide on sourcing strategy? Is there a way to come up with a customer-centric distribution strategy without analyzing individual shipping behavior of each of those customer stores? We will answer how to get around the details and arrive at strategies faster and reap benefits faster in this blog.
Network aggregation is a blanket term which includes product aggregation, customer aggregation and all other complexity reduction measures that are used while creating a network model. We will be talking about all the aforementioned aggregation methods and how umbrella strategies across markets, customer industry, product groups etc can be developed using the results to defining as well as the executing the goals of the company.
A network model with 10,000 products can be created but it will not help in achieving anything tangible. A properly done network model can solve many of a company’s problems. What do we mean by a proper network model? Something that is simple, recyclable and has high fidelity. Aggregation helps in achieving that. Aggregation is putting the data of your supply chain into logical groups for the purpose of modeling. Building and analyzing a network model is typically very data-intensive. Having thousands of similar nodes and tens of thousands of flows between them will make the model complex and create two problems:
Aggregation of nodes will reduce complexity for optimization algorithms when testing against real-world scenarios. This benefit will closely resound with the modeling community, since optimization run times can make a difference in productivity. Calculating optimal solutions more quickly means more scenarios can be tested in a given amount of time which means better responses to business questions and insights into supply chain operations.
Aggregation of nodes will also help in reducing the effort to decipher the results from the solve and will create a bird’s eye view of the network with the representative nodes we selected for aggregation. This helps in reducing the complexity at the same time retaining the fidelity of the network. For the reasons just discussed, reducing the quantity of nodes in the network model should be a priority when building the model. Let us look at some of the ways that we, at OPS Rules, aggregate data.
What are the most commonly used aggregation methods?
Product aggregation involves many different individual products to a small number of product families. In the example on the right hand side, we have 14 individual products aggregated into 3 different products [~80% reduction]. Although we know that they come under 2 different categories of sinks, we cannot just put them in 2 baskets. Why?
Product aggregation is dependent on other network parameters as much as it depends on products. The reason being the manufacturing locations of these products. In network design, source of your product plays an important role in developing strategies for the same. There is always an alternate sourcing opportunity available for a multi sourced product, whereas single sourced product is rigid as its name suggests. So putting them together will deprive the opportunity of alternate sourcing. Since some of the sinks of type B is produced in plant 1 and some others are produced in both plant 1 and plant 2, we have to separate them into 2 aggregated products.
This is one simple example of product aggregation where multiple factors affect the aggregation. To assure the fidelity of the network, checking at every turn of the aggregation process whether we are considering every factor that mayaffect the product. Making the final set of aggregated products as low as possible will make it easier for the optimizer to solve the model as well as to develop the umbrella strategies by product families and maintain the fidelity of the network will makes sure that those strategies work.
Customer aggregation involves many different actual delivery locations to several hundred aggregate points representing all the customers in a geographic area with other similar traits. Like Products, customers also have characteristics that enable them to be separated in groups. First level of aggregation will be based on these characteristics, The example below shows 16,000 Customers for an industrial manufacturer aggregated by 3-digit zip codes to 120 customers.
But again as with products, customers will also depend on other network parameters such as demand volume, market segment etc. We have to take into account all these characteristics before aggregating the customers so as to create a simple but high fidelity model.
Product and customer aggregation should be done on every network model unless the customer base or product variety is low. For complex large networks there are other types of aggregations that can be done in parallel with product and customer aggregation to cut the number of nodes to the required level. The less common types of aggregation are:
Supplier (from hundreds of vendors to a small group of important vendors or vendors in the same geographic area)
Plants, warehouses (that are a different building but on the same campus)
Time periods (modeling a year instead of every single day, Tactical Vs Strategic)
Cost types (from many different line items to a single cost figure)
How does aggregation help in defining strategies?
While the business strategy constitutes the overall direction that an organization wants to go, the supply chain strategy constitutes the actual operations of its supply chain to meet a specific objective. It will be an exercise in futility to come up with strategies for individual products, customer, plants etc. Aggregation helps in doing that. From the example we discussed in product aggregation, for the products from external supplier in china, we can choose to change the port of entry to reduce the transportation cost after analyzing the customer density for those products. For the dual sourced product from plant 1 and plant 2, we can choose to produce in plant 1 and close plant 2, thus reducing the overhead.
Similarly in the customer aggregation example, region focused strategies or market segment focused strategies can be developed once we have aggregated them in such groups. It helps in understanding the current network better [as in 5 states might have 70% of the demand or one market segment might cover most of the customers] and make better directional strategies [as to retain that market as well as making inroads into remaining regions]. Take the Channel map example above, Channel 3 is not showing up as much as Channel 1. Hence, for Channel 1 the strategy should be focused on how to retain that customer base but for Channel 3, the strategy should be focused on how to bring in new customers to the fold.
Also, we can aggregate products and customers based on demand. For high-demand products with low demand variability, the supply chain should have specialized and decentralized capacity. In contrast, for low-demand products with high variability, supply chain strategy should be made flexible with capacity that is centralized to pool demand.
Depending on the factors and methods we choose for aggregation, there are numerous aggregated parameters we can create to come up with strategies that fit each network. Aggregation is interesting because it is not only a natural thing to do when we think about the challenges being faced but also scary because we worry that we aren’t being accurate. It is imperative that aggregation be done for any network model, but note that more aggressive aggregation (fewer representative nodes) also increases the margin of error. There must be a balance between aggregation and the fidelity of the model. There are no hard and fast rules to do aggregation. What we discussed in this blog are few of the methods which worked for us. Data aggregation is more art than science. It depends on what you are trying to achieve from the model. Doing it right can have multiple benefits for your organization.
Accenture is a leading global professional services company, providing a broad range of services and solutions in strategy, consulting, digital, technology and operations. Combining unmatched experience and specialized skills across more than 40 industries and all business functions—underpinned by the world's largest delivery network—Accenture works at the intersection of business and technology to help clients improve their performance and create sustainable value for their stakeholders. With approximately 373,000 people serving clients in more than 120 countries, Accenture drives innovation to improve the way the world works and lives. Visit us at www.accenture.com.
Accenture Digital, comprised of Accenture Analytics, Accenture Interactive and Accenture Mobility, offers a comprehensive portfolio of business and technology services across digital marketing, mobility and analytics. From developing digital strategies to implementing digital technologies and running digital processes on their behalf, Accenture Digital helps clients leverage connected and mobile devices; extract insights from data using analytics; and enrich end-customer experiences and interactions, delivering tangible results from the virtual world and driving growth. Learn more about Accenture Digital at www.accenture.com/digital.
About Accenture Analytics
Accenture Analytics, part of Accenture Digital, delivers insight-driven outcomes at scale to help organizations improve their performance. With deep industry, functional, business process and technical experience, Accenture Analytics develops innovative consulting and outsourcing services for clients to help ensure they receive returns on their analytics investments. For more information follow us @ISpeakAnalytics and visit www.accenture.com/analytics.
This document makes descriptive reference to trademarks that may be owned by others. The use of such trademarks herein is not an assertion of ownership of such trademarks by Accenture and is not intended to represent or imply the existence of an association between Accenture and the lawful owners of such trademarks.
This blogpost is produced by consultants at Accenture as general guidance. It is not intended to provide specific advice on your circumstances. If you require advice or further details on any matters referred to, please contact your Accenture representative.